millenniumwaraigisfandomcom-20200223-history
Talk:Holy Warrior's Challenge/@comment-24.159.39.167-20161117221750/@comment-30176423-20161119045432
@A Fandom user * Yes, I just flipped a coin 6 times. Head, head, tail, head, tail, tail. I mean, what are the chances of that? Something like 1,6% I guess. Yes, exactly. You just decide beforehand that you are making a decision, which is based on this particular "head, head, tail, head, tail, tail" pattern. And then you make your coin flip experiment. If you get this particular result, then the coin is likely rigged. But if you don't get it, then you just can't reject the hypothesis about the coin being fair. Probably you just need to construct a different test if you want a bit more meaningful result? In our case, we are watching the total number of generated Maribel drops during the whole event (or during the initial part of it as the event progresses). Just because that's what is the most interesting thing during unit farming events. The theoretical probability of having no more than x''' drops in '''n runs where the probability of each drop is p''', can be calculated as '''binocdf(x, n, p) in the octave math application. And in a similar way, the theoretical probability of having at least x''' drops in '''n runs where the probability of each drop is p''', can be calculated as '''1 - binocdf(x - 1, n, p). Or you can use the online binomial calculator here: http://stattrek.com/online-calculator/binomial.aspx This is the only test that I'm relying on. And the decision to use it had been made before the Maribel event. If people get abnormally low or abnormally high number of drops compared to what is expected when using the constant drop probability model, then the constant drop probability model needs to be rejected. * By the way, the chance for a streak of 6 times hitting a 50% chance in a row within 39 tries is 25%. I never said you have only 6 tries to roll the dice. Who said that you can? Well, of course you can, but then you are not doing yourself a favour. Bragging about something that has 25% chance of happening does not make your case very unique, isn't it? On the other hand, if you look at the whole sequence, then you can do a test that is more statistically significant. * And if you start with 6 unlucky events and afterwards have 4 that perfectly match the 50:50 ratio, you will still end up believing in a 80% chance. Nope, this is just your imagination playing tricks. I'm using as many events as possible (preferably all events), or at least not cherry picking the convenient interval. * And when you roll 10 times the next day, you suddenly notices, that someone must have exchanged your dice, because the probabilties have changed. I'm just trying to make sure that I don't change my dice myself (even if they are from the same batch and supposedly identical). Maybe that's an unnecessary step. But it's really best to try to change as little as possible between the subsequent runs.